The role of cognitive load in navigational walking, planning and recall

Abstract

Walking is an important and common behavior in humans and is needed frequently in everyday life. The investigation of the influence of cognitive demands in different task difficulties on walking, more specifically on walking speed, was the aim of the first part of this thesis. Besides a decrease of walking speed with increasing cognitive demands it was also hypothesized that walking a known route will require less working memory resources than planning a route while walking. Therefore, a walking version of a Traveling Salesman task (planning while walking), as well as a walking version of a Corsi task (walking a known route) were designed. In both experiments the performance and walking speed decreased with increasing route and sequence length. There was no difference between the walking speeds in the two experiments but participants' performance was better in the Traveling Salesman task. Therefore, it is supposed that different working memory resources are required for solving the two tasks.

In the second part of this thesis it was investigated whether participants' performance in a Corsi task depends on the presentation and recalling type of the Corsi sequences. Besides the walking version of the Corsi task also a computerized version as well as two additional presentation and recall combinations were designed. Consequently, performance in four modality conditions was measured (i.e., Screen-Screen, Floor-Screen, Floor-Floor and Screen-Floor). Results revealed that in all experiments participants' performance decreased with increasing sequence length: This should be caused by the limited capacity of the working memory. Further, the findings indicate that different processes of spatial working memory are involved in the different modality conditions. For Screen-Screen only recall of the length of the sequence required working memory resources. For Floor-Screen and Floor-Floor additional demands on working memory were caused by reference frame transformation from floor to screen and spatial updating, respectively. In Screen-Floor all of these additional demands were required and therefore performance was poorest. It is suggested that performance in the Corsi task not only depends on the ability to recall the sequence but also on the additionally demands caused by the presentation and recall type.